OBJECTIVES: Commercial claims databases are thought to over-represent working age groups and under-represent older age groups because of employment bias attributable to their population of working age individuals, and their dependents, who have employer based insurance coverage. METHODS: Four disease classifications (acute coronary syndromes, dyslipidemia, breast cancer, and major depressive disorder[MDD]), and four drugs classes  (statins, anti-platelets, anti-estrogens and atypical antipsychotics) were used to develop and evaluate linear weighting methods compared to the United States (US) Census data. Any patient within the HealthCore Integrated Research Database (HIRD) with a claim for any of the disease or drug classifications during 2009 was eligible for this analysis. This year was selected because it was the latest year for which US Census data were available. Linear weighting method was used as the ‘Weight = Estimated Population Value from the US Census / HIRD value. RESULTS: A total of 14,794,609 patients in the HIRD met the inclusion criteria. Weights were calculated for the Mid-West, West, North-East and South geographic regions, and were stratified by gender and age using US Census age groups. For example,  for 1,227 North-East-based male patients, aged 15-19 years, in the HIRD database who were diagnosed with an MDD  the weight derived from US Census data was 1.10,  indicating that  an estimated 1,353 patients in the North-East region with MDD. CONCLUSIONS: The number of individuals in the US population with any one of four disease or drug classifications can be safely estimated using a linear weighting methodology, which makes it possible to calculate estimates of the total number of patients in the US population, or by geographic region, and further stratified by age group and gender. Such estimation is useful for research requiring estimation numbers of patients with a disease condition or pharmacy utilization.